SN
1999dn in NGC 7714 [Psc] SN
1999by in NGC 2841 [UMa] SN
2001bg in NGC 2608 [Cnc]
UK Supernova Discovery Team of Armstrong, Boles, Arbour, Foulkes and Laurie have discovered nearly three dozen SNe in the past five years. However positive confirmation of a supernova requires a spectrogram. This identifies certain features i.e. a broad silicon [SiII] absorption band centred at 612nm in orange light [type 1a] or hydrogen emission lines etc [type II]. Such spectra are usually obtained via professionally telescopes overseas.
With increasing and ever fainter
SNe being discovered - this service is declining and it may behold the
discoverer or the Team to do this work themselves. In the last
year work at WPO has been directed at a practical and economic solutions
to the problem. Recent WPO successes include confirmatory spectrograms
of Tom Boles' SN2001bg
and Mike Collins' Nova
Aql 2001 obtained with an efficient
transmission grating placed a centimetre or so from the detector.
These record both the SNe/Nova direct image and its spectrum offset to
the side - the separation of which allows direct measurement of spectral
lines - using a bright A type star like Vega for line calibration.
2001
Dec 9/10: Independently discovered
SN in NGC 7242 on 2001 Dec 7.8 by UK's Arbour/Armstrong at mag 15.3 - high
on meridian at dusk. Camera
rotated through 90o [north to right] to clear SN from galaxy
core. Numerous cumulative exposures on both evening [1m, 5m, 6m, 10m -
some co-added] through haze/fog show no obvious emission lines characteristic
of type II SN. Fairly confident of absorption line at 612nm
for SiII characteristic of type Ia SN shown adjacent overlaid with PNe/
nova/ SN2001bg spectra for comparison.
2001 Nov 22: This
evening a new fragile grating pellicle overlaid onto a clear IR block screw-in
filter for support some 40mm from CCD to yield an ultra low dispersion
of 14nm/pixel some x3 lower dispersion than Rainbow grating used
for most WPO non-objective slitless spectra. New grating = 40 lines/mm;
blazed 1st order; may reduce coma. The grating throughput is not as high
as the Rainbow [more light into zero order real images like m15/16
central stars in M57] but clearly resolves the emission lines in faint
WR stars and planetary nebulae like M57 and IC5117 at m13.5 in 60s exposures.
PNe type objects are needed to calibrate the spectrograph as narrow absorption
lines in stars like Vega will not be recorded. Hopefully the broad SiII
chasm
in type Ia SNe will be resolved. With tonight's haze and lvm~3.5
- this devise is promising for spectra of both I and II type SNe.
Relocating the grating onto the CCD window would lower dispersion to a remarkably low [but bright!] 50nm/pixel or x10 lower than Rainbow grating! Note: the filter mounted Rainbow grating cannot be mounted this close to the CCD.
2001 May 22/23: The latest results test a system where a transmission grating is in contact with and supported by the CCD window itself and partially overlays the CCD about 10mm* below. The slit-less devise is called a non-objective or field spectrograph and has a high throughput. It allows the two fields to be recorded separately - the galaxy field containing the suspect SNe with part of the CCD to the side allocated for a spectrum. This should present no obvious problems with the tiny size of the galaxy on the CCD and the larger CCDs used today. The telescope must be redirected to bring the suspect onto the spectrum field, focus is adjusted [focus does not coincide for both fields !] and a sufficiently long exposure of many minutes duration to record a satisfactory spectrogram.
* the effective focal length of the spectrograph.
The sample spectrograms and direct images below
obtained via a 300 l/mm grating**
and the small SX HX5 CCD [4.9mm x 3.6mm - 330 x 247 binned pixels] split
the CCD horizontally to satisfy the standard SX controlling software that
allows a line profile but only if parallel to the CCD long edge.
Other software like AstroArt can sample a spectrogram at any angle
across the CCD and the spectrum field could be allocated along the
short edge of the CCD as shown in the schematic. Spectra at
f/10 gave better images than at f/6.3 via a focal reducer. Ideally
the whole CCD camera assembly should be rotated so the suspect SNe and
the resultant spectrogram does not overlay field stars, the galaxy core
or other spectra. The Ha emission line
in Nova Aql 2001 is recorded in a brief 2m exposure.
**of relatively poor spectral throughput as
judged by the brightness of the zero order real image.
New pellix transmission grating
above proves much more efficient 1st order spectra.
text & images copyright Maurice Gavin - 2001